Example pipelines

You can always run your own pipeline scripts through the container, but the container also includes a set of predefined pipeline scripts that can be run as is or extended to your needs. Each pipeline script has a -h argument which will explain its use. The available pipelines are:

• preprocess-phix

• presto-abseq

• presto-clontech

• presto-clontech-umi

• changeo-10x

• changeo-igblast

• tigger-genotype

• shazam-threshold

• changeo-clone

All example pipeline scripts can be found in /usr/local/bin.

PhiX cleaning pipeline

Removes reads from a sequence file that align against the PhiX174 reference genome.

Usage: preprocess-phix [OPTIONS]

-s

FASTQ sequence file.

-r

Directory containing phiX174 reference db. Defaults to /usr/local/share/phix.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the input filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-h

This message.

Example: preprocess-phix

# Arguments
DATA_DIR=~/project
READS=/data/raw/sample.fastq
OUT_DIR=/data/presto/sample
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    preprocess-phix -s $READS -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    preprocess-phix -s $READS -o $OUT_DIR -p $NPROC

Note

The PhiX cleaning pipeline will convert the sequence headers to the pRESTO format. Thus, if the nophix output file is provided as input to the presto-abseq pipeline script you must pass the argument -x presto to presto-abseq, which will tell the script that the input headers are in pRESTO format (rather than the Illumina format).

NEBNext / AbSeq immune sequencing kit preprocessing pipeline

A start to finish pRESTO processing script for NEBNext / AbSeq immune sequencing data. An example for human BCR processing is shown below. Primer sequences are available from the Immcantation repository under protocols/AbSeq or inside the container under /usr/local/share/protocols/AbSeq. Mouse primers are not supplied. TCR V gene references can be specified with the flag -r /usr/local/share/igblast/fasta/imgt_human_tr_v.fasta.

Usage: presto-abseq [OPTIONS]

-1

Read 1 FASTQ sequence file. Sequence beginning with the C-region or J-segment).

-2

Read 2 FASTQ sequence file. Sequence beginning with the leader or V-segment).

-j

Read 1 FASTA primer sequences. Defaults to /usr/local/share/protocols/AbSeq/AbSeq_R1_Human_IG_Primers.fasta.

-v

Read 2 FASTA primer or template switch sequences. Defaults to /usr/local/share/protocols/AbSeq/AbSeq_R2_TS.fasta.

-c

C-region FASTA sequences for the C-region internal to the primer. If unspecified internal C-region alignment is not performed.

-r

V-segment reference file. Defaults to /usr/local/share/igblast/fasta/imgt_human_ig_v.fasta.

-y

YAML file providing description fields for report generation.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the read 1 filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-x

The mate-pair coordinate format of the raw data. Defaults to illumina.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-h

This message.

One of the requirements for generating the report at the end of the pRESTO pipeline is a YAML file containing information about the data and processing. Valid fields are shown in the example sample.yaml below, although no fields are strictly required:

sample.yaml

title: "pRESTO Report: CD27+ B cells from subject HD1"
author: "Your Name"
version: "0.5.4"
description: "Memory B cells (CD27+)."
sample: "HD1"
run: "ABC123"
date: "Today"

Example: presto-abseq

# Arguments
DATA_DIR=~/project
READS_R1=/data/raw/sample_R1.fastq
READS_R2=/data/raw/sample_R2.fastq
YAML=/data/sample.yaml
SAMPLE_NAME=sample
OUT_DIR=/data/presto/sample
NPROC=4

# Docker command
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    presto-abseq -1 $READS_R1 -2 $READS_R2 -y $YAML \
    -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    presto-abseq -1 $READS_R1 -2 $READS_R2 -y $YAML \
    -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

Takara Bio / Clontech SMARTer v1 immune sequencing kit preprocessing pipeline

A start to finish pRESTO processing script for Takara Bio / Clontech SMARTer v1 immune sequencing kit data. C-regions are assigned using the universal C-region primer sequences are available from the Immcantation repository under protocols/Universal or inside the container under /usr/local/share/protocols/Universal.

Usage: presto-clontech [OPTIONS]

-1

Read 1 FASTQ sequence file. Sequence beginning with the C-region.

-2

Read 2 FASTQ sequence file. Sequence beginning with the leader.

-j

C-region reference sequences (reverse complemented). Defaults to /usr/local/share/protocols/Universal/Mouse_IG_CRegion_RC.fasta.

-r

V-segment reference file. Defaults to /usr/local/share/igblast/fasta/imgt_mouse_ig_v.fasta.

-y

YAML file providing description fields for report generation.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the read 1 filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-x

The mate-pair coordinate format of the raw data. Defaults to illumina.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-h

This message.

Example: presto-clontech

# Arguments
DATA_DIR=~/project
READS_R1=/data/raw/sample_R1.fastq
READS_R2=/data/raw/sample_R2.fastq
CREGION=/usr/local/share/protocols/Universal/Human_IG_CRegion_RC.fasta
VREF=/usr/local/share/igblast/fasta/imgt_human_ig_v.fasta
SAMPLE_NAME=sample
OUT_DIR=/data/presto/sample
NPROC=4

# Docker command
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    presto-clontech -1 $READS_R1 -2 $READS_R2 -j $CREGION -r $VREF \
    -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    presto-clontech -1 $READS_R1 -2 $READS_R2 -j $CREGION -r $VREF \
    -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

Takara Bio / Clontech SMARTer v2 (UMI) immune sequencing kit preprocessing pipeline

A start to finish pRESTO processing script for Takara Bio / Clontech SMARTer v2 immune sequencing kit data that includes UMIs. C-regions are assigned using the universal C-region primer sequences are available from the Immcantation repository under protocols/Universal or inside the container under /usr/local/share/protocols/Universal.

Usage: presto-clontech-umi [OPTIONS]

-1

Read 1 FASTQ sequence file. Sequence beginning with the C-region.

-2

Read 2 FASTQ sequence file. Sequence beginning with the leader.

-j

C-region reference sequences (reverse complemented). Defaults to /usr/local/share/protocols/Universal/Human_IG_CRegion_RC.fasta.

-r

V-segment reference file. Defaults to /usr/local/share/igblast/fasta/imgt_human_ig_v.fasta.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the read 1 filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-x

The mate-pair coordinate format of the raw data. Defaults to illumina.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-a

Specify to run multiple alignment of barcode groups prior to consensus. This step is skipped by default.

-h

This message.

Example: presto-clontech-umi

# Arguments
DATA_DIR=~/project
READS_R1=/data/raw/sample_R1.fastq
READS_R2=/data/raw/sample_R2.fastq
CREGION=/usr/local/share/protocols/Universal/Human_IG_CRegion_RC.fasta
VREF=/usr/local/share/igblast/fasta/imgt_human_ig_v.fasta
SAMPLE_NAME=sample
OUT_DIR=/data/presto/sample
NPROC=4

# Docker command
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    presto-clontech-umi -1 $READS_R1 -2 $READS_R2 -j $CREGION -r $VREF \
    -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    presto-clontech-umi -1 $READS_R1 -2 $READS_R2 -j $CREGION -r $VREF \
    -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

10x Genomics V(D)J annotation pipeline

Assigns new annotations and infers clonal relationships to 10x Genomics single-cell V(D)J data output by Cell Ranger.

Usage: changeo-10x [OPTIONS]

-s

FASTA or FASTQ sequence file.

-a

10x Genomics cellranger-vdj contig annotation CSV file. Must corresponding with the FASTA/FASTQ input file (all, filtered or consensus).

-r

Directory containing IMGT-gapped reference germlines. Defaults to /usr/local/share/germlines/imgt/[species name]/vdj.

-g

Species name. One of human, mouse, rabbit, rat, or rhesus_monkey. Defaults to human.

-t

Receptor type. One of ig or tr. Defaults to ig.

-x

Distance threshold for clonal assignment. Specify “auto” for automatic detection. If unspecified, clonal assignment is not performed.

-m

Distance model for clonal assignment. Defaults to the nucleotide Hamming distance model (ham).

-e

Method to use for determining the optimal threshold. One of ‘gmm’ or ‘density’. Defaults to ‘density’.

-d

Curve fitting model. Applies only when method (-e) is ‘gmm’. One of ‘norm-norm’, ‘norm-gamma’, ‘gamma-norm’ and ‘gamma-gamma’. Defaults to ‘gamma-gamma’.

-u

Method to use for threshold selection. Applies only when method (-e) is ‘gmm’. One of ‘optimal’, ‘intersect’ and ‘user’. Defaults to ‘user’.

-b

IgBLAST IGDATA directory, which contains the IgBLAST database, optional_file and auxillary_data directories. Defaults to /usr/local/share/igblast.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the sequence filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-f

Output format. One of changeo or airr. Defaults to airr.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-i

Specify to allow partial alignments.

-z

Specify to disable cleaning and compression of temporary files.

-h

This message.

Example: changeo-10x

# Arguments
DATA_DIR=~/project
READS=/data/raw/sample_filtered_contig.fasta
ANNOTATIONS=/data/raw/sample_filtered_contig_annotations.csv
SAMPLE_NAME=sample
OUT_DIR=/data/changeo/sample
DIST=auto
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    changeo-10x -s $READS -a $ANNOTATIONS -x $DIST -n $SAMPLE_NAME \
    -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    changeo-10x -s $READS -a $ANNOTATIONS -x $DIST -n $SAMPLE_NAME \
    -o $OUT_DIR -p $NPROC

IgBLAST annotation pipeline

Performs V(D)J alignment using IgBLAST and post-processes the output into the Change-O data standard.

Usage: changeo-igblast [OPTIONS]

-s

FASTA or FASTQ sequence file.

-r

Directory containing IMGT-gapped reference germlines. Defaults to /usr/local/share/germlines/imgt/[species name]/vdj.

-g

Species name. One of human, mouse, rabbit, rat, or rhesus_monkey. Defaults to human.

-t

Receptor type. One of ig or tr. Defaults to ig.

-b

IgBLAST IGDATA directory, which contains the IgBLAST database, optional_file and auxillary_data directories. Defaults to /usr/local/share/igblast.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the sequence filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-f

Output format. One of airr (default) or changeo. Defaults to airr.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-k

Specify to filter the output to only productive/functional sequences.

-i

Specify to allow partial alignments.

-z

Specify to disable cleaning and compression of temporary files.

-h

This message.

Example: changeo-igblast

# Arguments
DATA_DIR=~/project
READS=/data/presto/sample/sample-final_collapse-unique_atleast-2.fastq
SAMPLE_NAME=sample
OUT_DIR=/data/changeo/sample
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    changeo-igblast -s $READS -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    changeo-igblast -s $READS -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

Genotyping pipeline

Infers V segment genotypes using TIgGER.

Usage: tigger-genotype [options]

-d DB, --db=DB

Change-O formatted TSV (TAB) file.

-r REF, --ref=REF

FASTA file containing IMGT-gapped V segment reference germlines. Defaults to /usr/local/share/germlines/imgt/human/vdj/imgt_human_IGHV.fasta.

-v VFIELD, --vfield=VFIELD

Name of the output field containing genotyped V assignments. Defaults to V_CALL_GENOTYPED.

-x MINSEQ, --minseq=MINSEQ

Minimum number of sequences in the mutation/coordinate range. Samples with insufficient sequences will be excluded. Defaults to 50.

-y MINGERM, --mingerm=MINGERM

Minimum number of sequences required to analyze a germline allele. Defaults to 200.

-n NAME, --name=NAME

Sample name or run identifier which will be used as the output file prefix. Defaults to a truncated version of the input filename.

-o OUTDIR, --outdir=OUTDIR

Output directory. Will be created if it does not exist. Defaults to the current working directory.

-f FORMAT, --format=FORMAT

File format. One of ‘airr’ (default) or ‘changeo’.

-p NPROC, --nproc=NPROC

Number of subprocesses for multiprocessing tools. Defaults to the available processing units.

-h, --help

Show this help message and exit

Example: tigger-genotype

# Arguments
DATA_DIR=~/project
DB=/data/changeo/sample/sample_db-pass.tab
SAMPLE_NAME=sample
OUT_DIR=/data/changeo/sample
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    tigger-genotype -d $DB -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    tigger-genotype -d $DB -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

TIgGER infers the subject-specific genotyped V gene calls and saves the corrected calls in a new column, v_call_genotyped. TIgGER also generates a *_genotype.fasta file, which contains the subject-specific germline IGHV genes. In future analyses, if v_call_genotyped column is used to replace v_call, please remember to use this *_genotype.fasta file generated previously by TIgGER as the subject-specific IGHV gene germline. An example of this application can be found in the Clonal assignment pipeline section.

Clonal threshold inference pipeline

Performs automated detection of the clonal assignment threshold.

Usage: shazam-threshold [options]

-d DB, --db=DB

Tabulated data file, in Change-O (TAB) or AIRR format (TSV).

-m METHOD, --method=METHOD

Threshold inferrence to use. One of gmm, density, or none. If none, the distance-to-nearest distribution is plotted without threshold detection. Defaults to density.

-n NAME, --name=NAME

Sample name or run identifier which will be used as the output file prefix. Defaults to a truncated version of the input filename.

-o OUTDIR, --outdir=OUTDIR

Output directory. Will be created if it does not exist. Defaults to the current working directory.

-f FORMAT, --format=FORMAT

File format. One of ‘airr’ (default) or ‘changeo’.

-p NPROC, --nproc=NPROC

Number of subprocesses for multiprocessing tools. Defaults to the available processing units.

--model=MODEL

Model to use for the gmm model. One of gamma-gamma, gamma-norm, norm-norm or norm-gamma. Defaults to gamma-gamma.

--cutoff=CUTOFF

Method to use for threshold selection. One of optimal, intersect or user. Defaults to optimal.

--spc=SPC

Specificity required for threshold selection. Applies only when method=’gmm’ and cutoff=’user’. Defaults to 0.995.

--subsample=SUBSAMPLE

Number of distances to downsample the data to before threshold calculation. By default, subsampling is not performed.

--repeats=REPEATS

Number of times to recalculate. Defaults to 1.

-h, --help

Show this help message and exit

Example: shazam-threshold

# Arguments
DATA_DIR=~/project
DB=/data/changeo/sample/sample_genotyped.tab
SAMPLE_NAME=sample
OUT_DIR=/data/changeo/sample
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    shazam-threshold -d $DB -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    shazam-threshold -d $DB -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

Clonal assignment pipeline

Assigns Ig sequences into clonally related lineages and builds full germline sequences.

If the TIgGER, or another package, was applied previously to the data set for identifying a subject-specific genotype, including potentially novel V, D and/or J genes, a new directory $NEW_REF with the personalized germline database should be created. For example, if TIgGER was run to identify a subject-specific IGHV genotype, the directory would contain: 1) *_genotype.fasta file generated previously by TIgGER, which contains the subject-specific germline IGHV genes 2) imgt_human_IGHD.fasta and imgt_human _IGHJ.fasta, which contain the IMGT IGHD and IGHJ genes and can both be copied from the original germline database: /usr/local/share/germlines/imgt/human/vdj/. When changeo-clone is called, this new personalized germline database should be passed with parameter -r (see example below). And please remember to update v_call column with subject-specific IGHV call (for TIgGER this is found in v_call_genotyped column).

# update v_call
db %>%
   dplyr::mutate(v_call = v_call_genotyped) %>%
   select(-v_call_genotyped)

Usage: changeo-clone [OPTIONS]

-d

Change-O formatted TSV (TAB) file.

-x

Distance threshold for clonal assignment.

-m

Distance model for clonal assignment. Defaults to the nucleotide Hamming distance model (ham).

-r

Directory containing IMGT-gapped reference germlines. Defaults to /usr/local/share/germlines/imgt/human/vdj.

-n

Sample identifier which will be used as the output file prefix. Defaults to a truncated version of the input filename.

-o

Output directory. Will be created if it does not exist. Defaults to a directory matching the sample identifier in the current working directory.

-f

Output format. One of airr (default) or changeo.

-p

Number of subprocesses for multiprocessing tools. Defaults to the available cores.

-a

Specify to clone the full data set. By default the data will be filtering to only productive/functional sequences.

-z

Specify to disable cleaning and compression of temporary files.

-h

This message.

Example: changeo-clone

# Arguments
DATA_DIR=~/project
DB=/data/changeo/sample/sample_genotyped.tab
DIST=0.15
SAMPLE_NAME=sample
OUT_DIR=/data/changeo/sample
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel \
    changeo-clone -d $DB -x $DIST -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif \
    changeo-clone -d $DB -x $DIST -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

Example: changeo-clone with personalized germline database

# Arguments
DATA_DIR=~/project
NEW_REF=/data/personalized_germlines
DB=/data/changeo/sample/sample_genotyped.tab
DIST=0.15
SAMPLE_NAME=sample
OUT_DIR=/data/changeo/sample
NPROC=4

# Run pipeline in docker image
docker run -v $DATA_DIR:/data:z immcantation/suite:devel changeo-clone -r $NEW_REF -d $DB -x $DIST -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC

# Singularity command
singularity exec -B $DATA_DIR:/data immcantation_suite-devel.sif changeo-clone -r $NEW_REF -d $DB -x $DIST -n $SAMPLE_NAME -o $OUT_DIR -p $NPROC